Exhaled Nitric Oxide and Clinical Phenotypes of Childhood Asthma

Bruno Mahut; Séverine Peyrard; Christophe Delclaux


Respiratory Research. 2011;12(65) 

In This Article


Between December 1, 2008 and April 14, 2010, 1001 NO measurements were performed in 592 asthmatic children (clinical diagnose) of which 398 had a positive bronchodilator response in their history. Of these 398 subjects, 169 met the criteria for inclusion in the cluster analysis (all the children fulfilling inclusion criteria were included). Non inclusion criteria were the absence of recent skin pricks test (n = 62), the absence of bronchodilator test on the day of visit (absence of treatment withdrawal, n = 130) and miscellaneous (n = 37). Table 1 shows the characteristics of the 169 asthmatic children.

First Approach

In univariate analysis, FENO was more elevated in atopic than in non atopic children (35[19–57] pbb versus 12 [9–16] ppb, p < 0.001), less elevated in controlled children (20 [13 #x2013; 37] ppb versus 34 [16–57] ppb, p = 0.018) and more elevated in children with ICS dose ≤ 200 μg/d (19 [13–34] versus 31 [16–57], p = 0.031). FENO correlated with height (r = 0.29, p < 0.001), age (r = 0.25, p = 0.001), FEV1 (r = 0.18, p = 0.021) and bronchodilator response (r = 0.26, p < 0.001). FENO was not decreased by tobacco exposure in univariate analysis. The multivariate analysis demonstrated that 4 variables independently correlated with log FENO (model: r = 0.60, p < 0.001): atopy (r = 0.46, p < 0.001), height (r = 0.29 p < 0.001), bronchodilator response (r = 0.26, p = 0.011) and ICS dose > 200 μg/d (r = −0.17, p = 0.019). Age, control and forced expiratory flows did not independently contribute to FENO variance.

Second Approach

Orthogonal varimax rotation was performed and the results are summarized in Table 2. Based on the pattern of loading we identified the factors as being representative of:

Factor 1: airway obstruction due to increased airway tonus (and airway to lung size ratio)

Factor 2: anthropometrics

Factor 3: treatment

Factor 4: severe exacerbation

Factor 5: atopy

Factor 6: FEV1 (airway remodeling)

Factor 7: symptoms

Factor 8: gender

Factor 9: tobacco smoke exposure

Communality of all variables was > 60% (excepting sRaw) and 74.8% of the total variance was explained by the factors (all factors had Eigenvalues > 1). The Kaiser-Meyer-Olkin measure of sampling adequacy was 0.643.

From this result we determined the 9 variables selected for cluster analyses (we favored continuous variables): gender, height, parental smoking, ICS dose, number of days with symptoms and requiring oral steroid, FEV1/FVC, FEV1 post BD and exhaled NO. Two additional variables were selected (1) FEF50%/TLC post-BD (index of airway/lung size) and (2) bronchodilator response (index of airway tonus).

A four-cluster model best fitted the population dataset, which were the following (Table 3): Cluster 1 (47%) described a subgroup of asthmatic boys, unexposed to tobacco, with well-controlled asthma, Cluster 2 (26%) described a subgroup of girls, unexposed to tobacco, with well-controlled asthma (similar to Cluster 1, excepting gender), Cluster 3 (6%) described a subgroup of girls or boys, unexposed to tobacco, with uncontrolled asthma associated with increased airway tone, and Cluster 4 (21%) also described a subgroup of girls or boys, exposed to parental smoking (either father, mother or both), with small airway to lung size ratio and uncontrolled asthma. The only difference (related to gender) between Cluster 1 and 2 was their airway to lung size ratio (p = 0.016, Fisher test).


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.